Home
http://www.risk.net/
Homeen-gbBacktesting expected shortfall: mission accomplished?A rigorous backtest for ES cannot exist, but a good approximation might do the job https://www.risk.net/our-take/6963251/backtesting-expected-shortfall-mission-accomplished
https://www.risk.net/our-take/6963251/backtesting-expected-shortfall-mission-accomplishedTue, 03 Sep 2019 16:18:00 +0100Podcast: Acerbi on backtesting ES and FRTB’s patchwork rulesBanque Pictet quant explains a new backtesting method for expected shortfall https://www.risk.net/risk-management/6950496/podcast-acerbi-on-backtesting-es-and-frtbs-patchwork-rules
https://www.risk.net/risk-management/6950496/podcast-acerbi-on-backtesting-es-and-frtbs-patchwork-rulesFri, 30 Aug 2019 04:30:00 +0100Podcast: McClelland on why you need a good MVA modelNumerix quant presents a model aimed at showing the total cost of a trade https://www.risk.net/cutting-edge/views/6886376/podcast-mcclelland-on-why-you-need-a-good-mva-model
https://www.risk.net/cutting-edge/views/6886376/podcast-mcclelland-on-why-you-need-a-good-mva-modelWed, 07 Aug 2019 04:30:00 +0100Fishing for collateral with neural netsSocGen quant uses deep learning technique to optimise collateral substitution https://www.risk.net/our-take/6875086/fishing-collateral-with-neural-nets
https://www.risk.net/our-take/6875086/fishing-collateral-with-neural-netsFri, 02 Aug 2019 14:40:30 +0100Podcast: Hans Buehler on deep hedging and the advantages of data-driven approachesQuant says a new machine learning technique could change the way banks hedge derivatives https://www.risk.net/derivatives/6705012/podcast-hans-buehler-on-deep-hedging-and-the-advantages-of-data-driven-approaches
https://www.risk.net/derivatives/6705012/podcast-hans-buehler-on-deep-hedging-and-the-advantages-of-data-driven-approachesWed, 05 Jun 2019 04:30:00 +0100Podcast: George Hong on quanto derivatives and Asia’s quant droughtCredit Suisse quant talks about new paper on valuing quanto options https://www.risk.net/derivatives/6609296/podcast-george-hong-on-quanto-derivatives-and-asias-quant-drought
https://www.risk.net/derivatives/6609296/podcast-george-hong-on-quanto-derivatives-and-asias-quant-droughtWed, 01 May 2019 04:30:00 +0100Podcast: Princeton’s Carmona on the future of quant educationCourse director discusses machine learning explainability and reclaiming game theory from economists https://www.risk.net/comment/6423241/podcast-princetons-carmona-on-the-future-of-quant-education
https://www.risk.net/comment/6423241/podcast-princetons-carmona-on-the-future-of-quant-educationMon, 25 Feb 2019 17:35:25 +0000Degree of influence: are machines starting to learn finance?This year's 'Degree of influence' analysis recognises a turning point in machine learning applications https://www.risk.net/cutting-edge/views/6212846/degree-of-influence-are-machines-starting-to-learn-finance
https://www.risk.net/cutting-edge/views/6212846/degree-of-influence-are-machines-starting-to-learn-financeFri, 14 Dec 2018 16:01:21 +0000Podcast: Mercurio on Libor, fraud and writing models on a planePost-Libor environment and financial crime detection to drive future research, says top quant https://www.risk.net/comment/5440081/podcast-mercurio-on-libor-fraud-and-writing-models-on-a-plane
https://www.risk.net/comment/5440081/podcast-mercurio-on-libor-fraud-and-writing-models-on-a-planeFri, 02 Mar 2018 12:15:14 +0000Intraday power storage and demand optionalityGeorge Levy discusses the value of intraday power storage and demand optionality in UK power contracts https://www.risk.net/cutting-edge/energy/5390016/intraday-power-storage-and-demand-optionality
https://www.risk.net/cutting-edge/energy/5390016/intraday-power-storage-and-demand-optionalityMon, 29 Jan 2018 04:30:00 +0000Podcast: Damiano Brigo on derivatives, AI, machine learning and moreGenuine artificial intelligence remains "very, very far away", says Imperial College's Brigo https://www.risk.net/comment/5389041/podcast-damiano-brigo-on-derivatives-ai-machine-learning-and-more
https://www.risk.net/comment/5389041/podcast-damiano-brigo-on-derivatives-ai-machine-learning-and-moreFri, 26 Jan 2018 04:30:00 +0000Degree of influence, 2017: Quants dissect initial marginInitial margin, optimal execution and applications of machine learning were the hottest topics of 2017 https://www.risk.net/cutting-edge/5374316/degree-of-influence-2017-quants-dissect-initial-margin
https://www.risk.net/cutting-edge/5374316/degree-of-influence-2017-quants-dissect-initial-marginWed, 13 Dec 2017 17:27:57 +0000Managing energy market volumetric riskKrzysztof Wolyniec presents a volumetric risk management model for energy markets https://www.risk.net/commodities/5092271/managing-energy-market-volumetric-risk
https://www.risk.net/commodities/5092271/managing-energy-market-volumetric-riskFri, 28 Apr 2017 14:46:22 +0100Optimal trading with linear and (small) non-linear costsBouchaud et al find the optimal trading strategy for a family of predictive signals in the presence of transaction costs https://www.risk.net/asset-management/3964451/optimal-trading-with-linear-and-small-non-linear-costs
https://www.risk.net/asset-management/3964451/optimal-trading-with-linear-and-small-non-linear-costsTue, 07 Mar 2017 04:00:00 +0000Derivatives funding, netting and accountingChristoph Burgard and Mats Kjaer expand their semi-replication framework to multiple counterparties https://www.risk.net/cutting-edge/banking/3964311/derivatives-funding-netting-and-accounting
https://www.risk.net/cutting-edge/banking/3964311/derivatives-funding-netting-and-accountingMon, 06 Mar 2017 13:55:00 +0000XVA at the exercise boundaryAndrew Green and Chris Kenyon show how the decision to exercise an option is influenced by XVAs https://www.risk.net/cutting-edge/banking/3847981/xva-at-the-exercise-boundary
https://www.risk.net/cutting-edge/banking/3847981/xva-at-the-exercise-boundaryFri, 03 Feb 2017 13:22:50 +0000Crunching mortality and life insurance portfolios with extended CreditRisk+Jonas Hirz, Uwe Schmock and Pavel Shevchenko present a summary of actuarial applications of the extended CreditRisk+
model https://www.risk.net/cutting-edge/investments/2479768/crunching-mortality-and-life-insurance-portfolios-with-extended-creditrisk
https://www.risk.net/cutting-edge/investments/2479768/crunching-mortality-and-life-insurance-portfolios-with-extended-creditriskFri, 06 Jan 2017 04:00:00 +0000Systemic risks in CCP networksBarker, Dickinson, Lipton and Virmani propose a credit and liquidity risk model for CCPs https://www.risk.net/cutting-edge/banking/2479766/systemic-risks-in-ccp-networks
https://www.risk.net/cutting-edge/banking/2479766/systemic-risks-in-ccp-networksThu, 05 Jan 2017 10:30:00 +0000A pairs trading strategy based on switching-regime volatility for commodity futuresA pairs trading strategy on energy, agricultural and index futures uses different parameters according to a volatility regime that is detected using a threshold evaluated in two ways, namely by means of a mixture of two Gaussian densities and a Markov-switching model. When associated to cointegration, this investment algorithm gives a larger Sharpe ratio with respect to classical methods https://www.risk.net/commodities/2479050/a-pairs-trading-strategy-based-on-switching-regime-volatility-for-commodity-futures
https://www.risk.net/commodities/2479050/a-pairs-trading-strategy-based-on-switching-regime-volatility-for-commodity-futuresTue, 06 Dec 2016 11:19:00 +0000Beat equal weighting: a strategy for portfolio optimisationYong (Jimmy) Jin and Lie Wang propose an estimation method for optimal portfolio weights under parameter uncertainty https://www.risk.net/cutting-edge/investments/2478967/beat-equal-weighting-a-strategy-for-portfolio-optimisation
https://www.risk.net/cutting-edge/investments/2478967/beat-equal-weighting-a-strategy-for-portfolio-optimisationMon, 05 Dec 2016 12:59:00 +0000Elasticity theory of structuringAndrei Soklakov presents a product design theory that incorporates Bayesian information processing and risk aversion https://www.risk.net/cutting-edge/banking/2478978/elasticity-theory-structuring
https://www.risk.net/cutting-edge/banking/2478978/elasticity-theory-structuringMon, 05 Dec 2016 12:45:00 +0000Commodity volatility, skew and inverse leverage effectTwo observations have consequences for commodity risk management and stochastic volatility modelling: the first is that the
standard leverage effects in commodities are due to a misspecification and are inefficient proxies for the forward slope effect; and the second is that commodity volatility can be driven exclusively by equity volatility over certain time periods. By Krzysztof Wolyniec https://www.risk.net/commodities/energy/2475855/commodity-volatility-skew-and-inverse-leverage-effect
https://www.risk.net/commodities/energy/2475855/commodity-volatility-skew-and-inverse-leverage-effectMon, 31 Oct 2016 17:03:27 +0000A profit and loss attribution framework for physical and financial energy portfoliosA profit and loss attribution framework can improve the information available to energy traders and give valuable insight
into the health of their portfolios https://www.risk.net/commodities/energy/2472025/a-profit-and-loss-attribution-framework-for-physical-and-financial-energy-portfolios
https://www.risk.net/commodities/energy/2472025/a-profit-and-loss-attribution-framework-for-physical-and-financial-energy-portfoliosTue, 27 Sep 2016 15:55:59 +0100Simulating meaningful uncertainty for complex energy portfoliosThe meaningful uncertainty simulation framework can extract all relevant information contained in market and portfolio data to give energy firms the ability to make informed business decisions https://www.risk.net/commodities/energy/2468523/simulating-meaningful-uncertainty-for-complex-energy-portfolios
https://www.risk.net/commodities/energy/2468523/simulating-meaningful-uncertainty-for-complex-energy-portfoliosMon, 22 Aug 2016 15:20:00 +0100How to get maximum value from power plant hedgingIt is common practice to delta hedge, ie re-hedge over time, a fuel-fired power plant’s production. Operators switch hedges from yearly to quarterly and monthly forwards as they become more liquid. Marco Miarelli and Margherita Grasso extend the existing literature on clean spark spread hedging and quantify the value added by the hedges’ refining technique https://www.risk.net/commodities/energy/2452501/how-get-maximum-value-power-plant-hedging
https://www.risk.net/commodities/energy/2452501/how-get-maximum-value-power-plant-hedgingFri, 25 Mar 2016 08:00:00 +0000Modelling the financial risks of wind generation with WeibullThis paper discusses the manner in which wind generation can affect the half-hourly APX price and also the risk
distributions associated with various power contracts. Wind generation is modelled using correlated doubly truncated
Weibull distributions, and analytic formulae for the values of puts and calls are presented https://www.risk.net/commodities/environment-renewables/2432786/modelling-the-financial-risks-of-wind-generation-with-weibull
https://www.risk.net/commodities/environment-renewables/2432786/modelling-the-financial-risks-of-wind-generation-with-weibullFri, 30 Oct 2015 13:18:00 +0000Quant ideas: Strategic versus tactical risk managementStrategic or enterprise risk management has long been seen as something of a Holy Grail for risk managers. However, the susceptibility of enterprise risk tools to poor quality data means the whole effort is likely to be misguided, argues Krzysztof Wolyniec, managing partner at Millwright Capital Management https://www.risk.net/commodities/energy/2428180/quant-ideas-strategic-versus-tactical-risk-management
https://www.risk.net/commodities/energy/2428180/quant-ideas-strategic-versus-tactical-risk-managementFri, 09 Oct 2015 08:00:00 +0100End-user hedging: FMC’s natural gas hedges show benefits of OTC tradingUsing a case study of Philadelphia-based FMC Corporation’s natural gas hedges, Ivilina Popova of Texas State University and Betty Simkins of Oklahoma State University show that over-the-counter derivatives are more efficient and effective compared with exchange-traded alternatives, and can reduce liquidity requirements https://www.risk.net/commodities/energy/2428211/end-user-hedging-fmcs-natural-gas-hedges-show-benefits-of-otc-trading
https://www.risk.net/commodities/energy/2428211/end-user-hedging-fmcs-natural-gas-hedges-show-benefits-of-otc-tradingFri, 02 Oct 2015 09:09:00 +0100Piterbarg leaves Barclays; bank names new head quantBarclays' head of quantitative analytics, Vladimir Piterbarg, has left the UK bank after 10 years to pursue other interests https://www.risk.net/people/2394311/piterbarg-leaves-barclays-bank-names-new-head-quant
https://www.risk.net/people/2394311/piterbarg-leaves-barclays-bank-names-new-head-quantMon, 09 Feb 2015 11:46:00 +0000Options for collateral optionsWhen collateral can be posted in multiple currencies, pricing even the simplest derivatives involves optionality, which is often tackled numerically. But by conditioning on a risk factor to make variables independent, this can be simplified. Alexandre Antonov and Vladimir Piterbarg show this is both quicker and more accurate than more obvious methods https://www.risk.net/derivatives/2328971/options-collateral-options
https://www.risk.net/derivatives/2328971/options-collateral-optionsMon, 17 Feb 2014 10:44:00 +0000The simple link from default to LGDRisk managers often use ad hoc regressions to incorporate loss given default risk into their models. Regression can give good results when data is plentiful, but in reality this data is sparse. Here, Jon Frye introduces a new function that can be more accurate and is easy to implement https://www.risk.net/risk-management/credit-risk/2328812/simple-link-default-lgd
https://www.risk.net/risk-management/credit-risk/2328812/simple-link-default-lgdFri, 14 Feb 2014 10:43:00 +0000Local correlation familiesLocal correlation models allow for better hedging by incorporating changes in co-dependence in the delta. Using the particle method and affine transforms, Julien Guyon builds families of local correlation models that calibrate to the market smile of a basket and allow for a view on correlation skew or other properties
https://www.risk.net/derivatives/2325805/local-correlation-families
https://www.risk.net/derivatives/2325805/local-correlation-familiesWed, 29 Jan 2014 15:29:00 +0000SABR symmetryTypical implementations of the stochastic alpha beta rho model involve asymptotic expansion approximations, which can generate inaccurate prices for long-dated options. But directly solving a pricing partial differential equation incurs high computational costs. Hyukjae Park shows how the model’s symmetry can be harnessed to reduce the complexity of the calculation and improve accuracy over expansions https://www.risk.net/derivatives/2322292/sabr-symmetry
https://www.risk.net/derivatives/2322292/sabr-symmetryFri, 10 Jan 2014 14:24:00 +0000Differential rates, differential pricesCollateral agreements and funding costs affect derivatives prices through discounting and adjustments. But if the borrowing and lending rates aren’t equal, the situation becomes even more complicated. Fabio Mercurio shows that buy and sell prices diverge, and the resulting non-linear valuation means a portfolio may not be the sum of its parts https://www.risk.net/derivatives/2321764/differential-rates-differential-prices
https://www.risk.net/derivatives/2321764/differential-rates-differential-pricesWed, 08 Jan 2014 18:07:00 +0000Funding strategies, funding costsThe economic value of derivatives is influenced by funding costs, because the costs imply windfalls or shortfalls to bondholders on a bank’s default. But the resulting adjustments depend not just on the funding spread but on the funding strategy deployed. It is another layer of complexity to derivatives pricing. By Christoph Burgard and Mats Kjaer https://www.risk.net/derivatives/2310207/funding-strategies-funding-costs
https://www.risk.net/derivatives/2310207/funding-strategies-funding-costsFri, 29 Nov 2013 19:00:00 +0000Systematic risk factors redefinedCredit risk factor models tend to have a narrow focus on the Gaussian case, use copula functions that don’t work well with the martingale methods used in pricing, and can introduce arbitrage. Dariusz Gatarek and Juliusz Jablecki show how an increasing sequence of default times can be used to create systematic factors that allow for a rich correlation structure – and keep strong links with pricing https://www.risk.net/risk-management/credit-risk/2303417/systematic-risk-factors-redefined
https://www.risk.net/risk-management/credit-risk/2303417/systematic-risk-factors-redefinedMon, 28 Oct 2013 15:08:00 +0000Stuck with collateralValues of derivatives contracts for which there is a choice of asset to post as collateral have been shown to depend on the instantaneous interest rates these assets return. However, the models developed so far assumed perfect substitution. In reality, such agreements are hard to enforce. Vladimir Piterbarg shows that in that case, the optimal posting strategy is actually determined by a suitable term collateral rate https://www.risk.net/derivatives/2302926/stuck-with-collateral
https://www.risk.net/derivatives/2302926/stuck-with-collateralFri, 25 Oct 2013 12:17:00 +0100Pricing CDSs’ capital reliefPositions in credit default swaps (CDSs) are eligible instruments to reduce some Basel III capital requirements. The value of this benefit should be reflected in the price. Chris Kenyon and Andrew Green incorporate this into a pricing model for CDSs, and show it may account for more than half the spread https://www.risk.net/derivatives/credit-derivatives/2296602/pricing-cdss-capital-relief
https://www.risk.net/derivatives/credit-derivatives/2296602/pricing-cdss-capital-reliefFri, 27 Sep 2013 10:58:00 +0100Hedge backtesting for model validationDerivatives pricing and expected exposure models must be backtested as a basic regulatory requirement. But what does this mean exactly, and how can it be used to reserve against model risk? Lee Jackson introduces a general backtesting framework for market-calibrated models, making the link with financial theory, and shows how it can inform recalibration and help insulate banks from model failure https://www.risk.net/risk-management/2295754/hedge-backtesting-model-validation
https://www.risk.net/risk-management/2295754/hedge-backtesting-model-validationFri, 20 Sep 2013 16:03:00 +0100Exposure under systemic impactWrong-way risk (WWR) behaves differently for exposures to systemically important counterparties because their default has the potential to move financial markets before the close-out. Michael Pykhtin and Alexander Sokol show how the traditional exposure simulation framework can be adapted to account for this systemic WWR. They illustrate how the model can be calibrated by using market credit default swap spreads and historical defaults – including those of Russia, Argentina and Lehman Brothers https://www.risk.net/risk-management/credit-risk/2290019/exposure-under-systemic-impact
https://www.risk.net/risk-management/credit-risk/2290019/exposure-under-systemic-impactTue, 20 Aug 2013 15:41:00 +0100Fast gammas for Bermudan swaptionsAdjoint differentiation is an efficient way to accurately calculate the Greeks of Libor derivatives by Monte Carlo simulation. Ralf Korn and Qian Liang extend this to calculate the gamma matrices of Bermudan swaptions more quickly than existing approaches https://www.risk.net/derivatives/interest-rate-derivatives/2289183/fast-gammas-bermudan-swaptions
https://www.risk.net/derivatives/interest-rate-derivatives/2289183/fast-gammas-bermudan-swaptionsThu, 15 Aug 2013 15:12:00 +0100Hybrid smiles made fastRisk management of equity-linked structured notes requires consistent modelling of both stocks’ smiles and stochastic interest rates. Existing approaches require costly computations to capture highly curved smiles – especially at long-dated maturities. Messaoud Chibane and Dikman Law show how a quadratic parameterisation of volatility, with some analytic approximations, can be much quicker https://www.risk.net/derivatives/structured-products/2284871/hybrid-smiles-made-fast
https://www.risk.net/derivatives/structured-products/2284871/hybrid-smiles-made-fastThu, 25 Jul 2013 15:45:00 +0100SABR spreads its wingsTraditional methods for the stochastic alpha beta rho model tend to focus on expansion approximations that are inaccurate in the long maturity ‘wings’. However, if the Brownian motions driving the forward and its volatility are uncorrelated, option prices are analytically tractable. In the correlated case, model parameters can be mapped to a mimicking uncorrelated model for accurate option pricing. Alexander Antonov, Michael Konikov and Michael Spector explain how https://www.risk.net/derivatives/interest-rate-derivatives/2284866/sabr-spreads-its-wings
https://www.risk.net/derivatives/interest-rate-derivatives/2284866/sabr-spreads-its-wingsThu, 25 Jul 2013 15:41:00 +0100A quadratic volatility Cheyette modelQuasi-Gaussian or Cheyette interest rate models provide derivatives desks with solutions to some of the Libor market model’s problems in an explicitly Markovian representation. Here, Messaoud Chibane and Dikman Law introduce a local volatility extension and an efficient calibration scheme https://www.risk.net/derivatives/interest-rate-derivatives/2277261/quadratic-volatility-cheyette-model
https://www.risk.net/derivatives/interest-rate-derivatives/2277261/quadratic-volatility-cheyette-modelTue, 25 Jun 2013 16:42:00 +0100Smile in the low momentsSkew and curvature of volatility smiles are not only difficult to estimate, but also poorly reproduced by most smile expansions. Jean-Philippe Bouchaud, Lorenzo De Leo, Vincent Vargas and Stefano Ciliberti propose an expansion that effectively captures these dynamics by interpreting its parameters as payouts of exotic options for which efficient pricing methods are readily available https://www.risk.net/derivatives/equity-derivatives/2277256/smile-low-moments
https://www.risk.net/derivatives/equity-derivatives/2277256/smile-low-momentsTue, 25 Jun 2013 16:26:00 +0100SABR goes normalThe benchmark stochastic alpha beta rho model for interest rate derivatives was designed for an environment of 5% base rates, but its traditional implementation method based on a lognormal volatility expansion breaks down in today’s low-rate and high-volatility environment, returning nonsensical negative probabilities and arbitrage. Philippe Balland and Quan Tran present a new method based on a normal volatility expansion with absorption at zero, which calibrates while eliminating arbitrage in the lower strike wing https://www.risk.net/derivatives/interest-rate-derivatives/2269613/sabr-goes-normal
https://www.risk.net/derivatives/interest-rate-derivatives/2269613/sabr-goes-normalTue, 28 May 2013 15:08:00 +0100Lois: credit and liquidityThe spread between Libor and overnight index swap rates used to be negligible – until the crisis. Its behaviour since can be explained theoretically and empirically by a model driven by typical lenders’ liquidity and typical borrowers’ credit risk. By Stéphane Crépey and Raphaël Douady https://www.risk.net/derivatives/2269615/lois-credit-and-liquidity
https://www.risk.net/derivatives/2269615/lois-credit-and-liquidityTue, 28 May 2013 14:59:00 +0100Collateral convexity complexityThe crisis abolished the risk-free rate, and brought the role of credit support annexes to the fore in derivatives pricing. Paul McCloud develops the general pricing framework that allows the convexity effects to be captured https://www.risk.net/derivatives/2257630/collateral-convexity-complexity
https://www.risk.net/derivatives/2257630/collateral-convexity-complexityFri, 26 Apr 2013 16:13:00 +0100CDSs, CVA and DVA – a structural approachCounterparty risk is difficult to include systematically in credit default swap pricing. Reviving Merton’s structural approach – and generalising to higher dimensions – makes it tractable. By Alex Lipton and Ioana Savescu https://www.risk.net/derivatives/credit-derivatives/2256702/cdss-cva-and-dva-structural-approach
https://www.risk.net/derivatives/credit-derivatives/2256702/cdss-cva-and-dva-structural-approachFri, 22 Mar 2013 16:46:00 +0000Breaking break clausesThe value of early termination clauses in derivatives depends crucially on the type of close-out value used and on the counterparty risk, and embeds optionality in even the most vanilla swap contracts. In the case of the so-called risk-free close-out, a deterministic default intensity model can be used to price them. By Lorenzo Giada and Claudio Nordio https://www.risk.net/derivatives/2256797/breaking-break-clauses
https://www.risk.net/derivatives/2256797/breaking-break-clausesFri, 22 Mar 2013 16:42:00 +0000